Reference Equations for Assessing the Physical Activity of Japanese Patients with Chronic Obstructive Pulmonary Disease

Abstract

Background: To improve physical activity (PA) in patients with chronic obstructive pulmonary disease (COPD), providing a target PA value based on the individual patient's condition may be a useful interventional strategy. However, to determine the target value, a predictive PA value for each patient is required.

Research question: What is the reference equation consisting of PA-related factors to determine the predictive PA value for each patient with COPD?

Material and methods: In this prospective cross-sectional observational study, we measured the PA with a triaxial accelerometer and several other factors including demographic factors, pulmonary function, dyspnea, exercise capacity, muscle strength, nutrition, and indicators of several comorbidities in stable Japanese outpatients with COPD aged ≥40 years old and detected PA-related factors by a multiple regression analysis and stepwise method. We created reference equations for four indices of PA using multiple linear regression equations.

Results: Two hundred and twenty-seven patients were registered. The equations of duration at ≥2.0 metabolic equivalents (METs) and step count consisted of 4 factors: 6-minute walk distance, modified Medical Research Council dyspnea scale, anxiety score of the Hospital Anxiety and Depression Scale, and the forced expiratory volume in 1 second % of predicted value. Those of duration at ≥3.0 METs and total activity at ≥3.0 METs consisted of 5 factors: the above 4 factors and age or brain natriuretic peptide. There was no fixed bias or proportional bias between the measured and predictive values in patients with non-high measured PA values.

Conclusion: We determined reference equations for four indicators of PA using PA-related factors in Japanese patients with COPD. The predictive values calculated using the equations could be useful for deciding target PA values for each patient.

Clinical trial registration: UMIN-CTR; No.: UMIN000025459; URL: https://www.umin.ac.jp/ctr/index.htm.

Keywords: COPD; metabolic equivalents; predictive value; step count.

Figure 1

Flow diagram.

Figure 2

Reference equations of physical activity. METs, metabolic equivalents. Units: Duration at ≥2.0 METs (min), Duration at ≥3.0 METs (min), Total activity at ≥3.0 METs (METs∙h), Step count (steps), 6MWD (m), mMRC (point), HADS-A (point), FEV1 %pred (%), age (years old), BNP (pg/mL).

Figure 3

Relationships between the measured and predictive values. (A) duration at ≥2.0 METs; r=0.462, P<0.0001, (B) duration at ≥3.0 METs; r=0.598, P<0.0001, (C) total activity at ≥3.0 METs; r=0.614, P<0.0001, (D) Step count; r=0.539, P<0.0001.

Figure 4

Bland-Altman Plots. (A) duration at ≥2.0 METs, (B) duration at ≥3.0 METs, (C) total activity at ≥3.0 METs, (D) Step count.

Figure 5

Bland-Altman Plots after excluding highly active patients. (A) duration at ≥2.0 METs (patients with <231 minutes of measured value), (B) duration at ≥3.0 METs (patients with <87 minutes of measured value), (C) total activity at ≥3.0 METs (patients with <4.4 METs∙h of measured value), (D) Step count (patients with <5108 steps of measured value).